The result of electron microscopic investigations of gas-vacuoles in a culture of the benthal alga Oscillatoria chalybea was compared with the extensive literature concerning gas-vacuole formation and virus infection in bacteria and animals.On the basis of 21 features indicating that gas-vacuoles are pathological inclusions it was concluded that they are viruslike particles.Viruslike structures have also been found in our material of Symploca muscorum, Plectonema boryanum, a Nostoc sp., and a new case of virus infection in the thermopldlic alga Mastigocladus laminosus has been described.In a previous paper the author [15] pointed out that light microscopic features such as formation of calyptra, constrictions at the cross walls, cyanophycin granulesoften arranged at the cross walls -, sheath formation, and the direction in whichliving trichomes turn, are features which are without taxonomic value. Further it says in the paper that another problem to be solved is the associate bacteria, fungi, and viruses and their influence on the habitual features of the blue-green algae. So far, we know very little about the morphological effect of such associates, but there are probably many morphological features which might be explained if we understood the influence of associated microorganisms, cp. also FOREST [18].During the last few years the author, therefore, has also used electron microscopes for his investigations of blue-green algae. These investigations have confirmed what alredy GEITLER [20] wrote in his manual of Cyanophyceae that vacuole formations in the blue-green algae amount to a pathological change, and it should be added that this change may be caused by bacteria, fungi or viruses.Quite a number of the blue-green algae shown in literature are clearly marked by the effect of this pathological influence. In the cases where these pathological features are not part of the diagnosis of the species, they are of no importance for the taxonomy, but it must be pointed out that the pathological changes are in many cases included in and form an essential part of the diagnosis of the species.It appears evident that a pathological change of features cannot ]ustify'a separate taxon.